Create app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import requests
|
| 2 |
+
from bs4 import BeautifulSoup
|
| 3 |
+
|
| 4 |
+
url = "https://aspireec.com/"
|
| 5 |
+
response = requests.get(url)
|
| 6 |
+
soup = BeautifulSoup(response.text, 'html.parser')
|
| 7 |
+
# Extract data (e.g., headlines, paragraphs, etc.)
|
| 8 |
+
content = soup.find_all('p') # Example: extracting paragraphs
|
| 9 |
+
website_data = [p.text for p in content]
|
| 10 |
+
import json
|
| 11 |
+
|
| 12 |
+
with open('website_data.json', 'w') as file:
|
| 13 |
+
json.dump(website_data, file)
|
| 14 |
+
from sentence_transformers import SentenceTransformer
|
| 15 |
+
|
| 16 |
+
model = SentenceTransformer('all-MiniLM-L6-v2')
|
| 17 |
+
embeddings = model.encode(website_data)
|
| 18 |
+
import faiss
|
| 19 |
+
import numpy as np
|
| 20 |
+
|
| 21 |
+
# Create FAISS index
|
| 22 |
+
dimension = embeddings.shape[1]
|
| 23 |
+
index = faiss.IndexFlatL2(dimension)
|
| 24 |
+
index.add(np.array(embeddings))
|
| 25 |
+
query = "What is the website about?"
|
| 26 |
+
query_embedding = model.encode([query])
|
| 27 |
+
distances, indices = index.search(np.array(query_embedding), k=1)
|
| 28 |
+
best_match = website_data[indices[0][0]]
|
| 29 |
+
from transformers import pipeline
|
| 30 |
+
|
| 31 |
+
summarizer = pipeline("summarization", model="google/flan-t5-base")
|
| 32 |
+
answer = summarizer(best_match)
|
| 33 |
+
print(answer)
|
| 34 |
+
import streamlit as st
|
| 35 |
+
|
| 36 |
+
st.title("Website Chatbot")
|
| 37 |
+
|
| 38 |
+
user_input = st.text_input("Ask me anything about the website:")
|
| 39 |
+
if user_input:
|
| 40 |
+
response = get_answer(user_input) # Function to query data
|
| 41 |
+
st.write(response)
|